CN102393840B - Entity detection and extraction for entity cards - Google Patents
Entity detection and extraction for entity cards Download PDFInfo
- Publication number
- CN102393840B CN102393840B CN201110164789.0A CN201110164789A CN102393840B CN 102393840 B CN102393840 B CN 102393840B CN 201110164789 A CN201110164789 A CN 201110164789A CN 102393840 B CN102393840 B CN 102393840B
- Authority
- CN
- China
- Prior art keywords
- entity
- response results
- classification
- class
- entities
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
Abstract
A system and method are provided for detecting entity information contained within search results. The detected entity information can be used to determine a category of entity as well as a specific entity within the search results. Entity information can be extracted from the documents associated with the search results. This information can be used as part of the information for an entity card, which can be displayed to a user in conjunction with and/or in place of the search results.
Description
Technical field
The present invention relates to search inquiry technology in networking technology area, more particularly, to network technology.
Background technology
Search engine is used for searching various types of information.Although lists of links is returned to relevant documentation such as
The present is a kind of familiar form, but this not necessarily convenient form.In order to find specific certain information,
User's typically required click-through chain fetches checks corresponding document.If information needed do not accessed in user
In one document, then user may need this process is repeated several times.
Content of the invention
In embodiments, there is provided be included in the system of entity information and the side in Search Results for detection
Method.Entity class and the concrete entity in Search Results can be determined using the entity information being detected.Can
Entity information is extracted from the document being associated with Search Results.This information is used as the information of entity card
A part, can display to the user that this information combine Search Results and/or replace Search Results.
There is provided present invention will in detailed description below further to introduce in simplified form
Some concepts of description.This content of the invention is not intended to identify key feature or the essence of theme required for protection
Feature, is not intended to the scope being exploited separately for assisting in theme required for protection.
Brief description
Describe the present invention below with reference to the accompanying drawings in detail, wherein:
Fig. 1 applies to realize the block diagram of the exemplary computing environments of various embodiments of the present invention.
Fig. 2 schematically shows the example of the system being adapted for carrying out embodiments of the invention.
The flow chart that Fig. 3 depicts method according to an embodiment of the invention.
The flow chart that Fig. 4 depicts method according to an embodiment of the invention.
The flow chart that Fig. 5 depicts method according to an embodiment of the invention.
Specific embodiment
General view
In embodiments, when receiving search inquiry, search engine can generate multiple Search Results.Subsequently
The result being generated by search engine can be analyzed, to identify the entity class being indicated by this result.This mark can
It is based in part on the mark of one or more websites towards classification in this result.This knot can be analyzed further
Really, to determine intention entity.Based on be intended to entity, be ready for corresponding to this entity entity card and with search
Result shows together.Optionally, can be excluded in the Search Results being generated based on expectation entity
Individual or multiple display or be incorporated in entity card.
In the following discussion, entity card refers to that the information of enhanced entity-specific presents.Entity card may include pass
Various types of information in entity.Entity card can allow to assume these letters to user in response to search inquiry
Breath is not so that user needs to screen document links to obtain this information.
Class template
Determine that the user view being associated with search inquiry can cause various problems.One kind is used for identifying user's meaning
The method of figure can be to determine whether search inquiry is related to entity.Entity may refer to such as author, politics
The individual of the type such as family or athlete；The product of the types such as film, book or consumer goodss；Or such as eat
The place of the types such as the Room, hotel, the public place of entertainment or retail shop.However, the mark reality related to search inquiry
Body also result in difficulty.Many conventional methods attempt to set up the entity that can be mated with the item in search inquiry
List.It is probably difficult and time-consuming that these lists are kept up-to-date.In addition, it is related to search inquiry
Entity may be not included in search terms.
In embodiments, entity information can be based on the Search Results in response to search inquiry come dynamically true
Fixed.The Search Results that may be based partly on mark from known document corresponding with particular category to identify reality
Body.There are the many websites attempting to follow the tracks of the current state of various entities.For example, follow the tracks of film, hotel,
Multiple web positions of consumption electronic product or book are available.These websites are referred to alternatively as the station towards classification
Point.Current development towards in the website generally tracking specific category interested of classification, and can thus provide
Current information with regard to the entity in classification.Generally delay in time towards the quantity of the website of classification and/or identity
Slow change, is possibly manageable task related to classification therefore by suitable site identity.With unified money
Source location symbol (url) is associated and is associated with classification from one of these websites document of website
Probability may increase.
For the document from the website towards classification, one or more class templates can be constructed.Be located at towards
It is often consistent between the described entity on website of the file structure of the website of classification.Can use and present
Concordance come to be configured to slave site extract information template.For example, the face of the information with regard to film is provided
Generally consistent assume form to the website of classification by having.Movie director will be marked by ad hoc fashion, all
Such as mark in a document specific local or using title adjacent with director names and/or in the above
" director " is marking.Can be configured to extract the template of information from document using the desired form that presents.
Note, website can be considered as the website more than a classification towards type.For example, online retailer can
Sell and include the products such as consumption electronic product, dvd and computer game.This online retailer can have
One or more url assemblies corresponding with each field in these fields.Thus, look into depending on search
Ask, the outward appearance from the document of online retailer can be corresponding with movies category, game classification or consumer goodss classification.
Can be each construction template of the website towards classification.Template may include at least two assemblies.Template
A part can be url assembly.Url assembly represents the initial part of url.With url template just
The document of initial portion coupling can be from the document of the known website towards classification.Second assembly of template
Can be to extract format component.Extracting format component provides the specification of multiple data fields, every including being directed to
The type of the information that one data field extracts, and the specification how extracting this information.Any convenience can be used
Type specification.For example, specification can identify the particular location in document, to retrieve information, such as from literary composition
The second field in shelves fifth line obtains a value.Or, specification can be that label drives, and such as specifies first
The head of mark such as " title " or " movie title ", and then take that and look there is certain with head
Related information or word.
In addition to based on the class template towards the website of classification, can have open form for category construction
One or more class templates.Open form class template can be constructed, to extract and the website towards classification
Template identical information.The template phase that open format module can be driven with the label of the website towards classification
Seemingly, because open format module will be applied to the unmatched page with url assembly.
Note, can by each open format module be applied to each response results or be applied to be identified as with
The corresponding each response results of entity being identified.This may result in and extracts each data field from identical document
Multiple values.In order that this data is more useful to each document, which executable consistency detection is to determine
Open format module is successful for the data of extraction data-oriented field.For example, for given document,
Multiple values of each field can be made comparisons with the value extracted from the document from the website towards classification.Because
The probability of accidentally coupling is relatively low, so matching value is likely to be the value of correct extraction.Another type of inspection
It can be the consistency check to the value extracted using the open format module from other documents.Equally,
The probability of accidentally coupling is relatively low, and therefore coupling is possible to indicate the successful extraction of this field.
The website towards classification can be determined by any convenient method.Can manually identify towards classification
Website.Or, can be determined towards classification by submitting the known search that should return the special result of classification to
Website.The website most frequently occurring can be considered as the website towards classification.
Classification and entity identification
When receiving search inquiry, multiple response results or document can be generated using traditional search engines.
In following embodiment, can analyze response document a part of determining classification or entity information.This part
Can be corresponding with the response results of first 10 or front 20 or front 50 or any conventional amounts.Can divide
Analysis response document is to determine entity class.The a part of of analysis can be by the url of document and class template
Assembly is mated.It may desirable, in one embodiment at least one url component matching, to enter
The mark of row entity class.Another part of analysis can be by from the metadata of Search Results and known terms
Mated.For example, can be by metadata items such as " films ", " retailer " or " film " and electricity
Shadow website is associated.Metadata can with the metatag of document or with the document to show as Search Results
Captions or when document to return as Search Results available and any other information of being associated with document
Corresponding.
The coupling of class template or metadata can be subsequently weighted, with determine for search inquiry whether with class
Not corresponding fraction.For example, the fraction of the category can be made with each document of url component matching
Contribution.Extra weight or fraction can be distributed to the first document with url component matching.With respect to
Extra weight or fraction can be distributed to ranking higher by the relatively low Search Results of the ranking of url component matching
Search Results.The weighting of similar type can be used for metadata analysis.
Can determine that the intention classification of search based on fraction.For example, if detecting three of single classification or more
Many url assembly, then can distribute to the category by inquiry.Detect if based on the coupling of url assembly
Multiple classifications, then can distribute the classification of top ranked.In certain embodiments, without url is detected
, then may not there is the selection of classification in component matching.Or, if there is one or less url group
Part mates, then can not carry out the selection of classification.
Also can analysis result to determine whether entity is associated with search inquiry.In one embodiment, can be first
First identified category, and subsequently can analysis result to determine entity.In this embodiment, only consider to belong to institute
The entity of the classification of mark.In another embodiment, without entity class is detected, then do not have and search
The associated entity of rope inquiry.
The a part of of entity analysis can be the document being applied to class template from the website towards classification.
Because document is derived from the website towards classification, so the extraction form of document is likely to be known.Thus,
It is possible to a part for document corresponding with entity it could also be possible that known, and entity can directly be carried
Take.Another part of entity analysis can be to be applied to respond by the one or more of open form class template
It is not from the document of the website towards classification in result.For example, dining room name is listed in many dining rooms comment website
Claim together with address.Open format module will attempt by finding one group text corresponding with address come from unknown literary composition
Dining room title is extracted in shelves form.Subsequently can extract the name before address and be referred to as possible entity.
In each embodiment that analysis files in open formats is not determined with classification before detecting entity, used opens
Put format module can website towards classification any with Search Results classification corresponding.
The solid data extracting from document subsequently can be analyzed, with the entity that determination is associated with search inquiry be
No can be identified.This analysis may compare extracted information, with determine whether to only exist a possible entity,
Or whether can select an entity from some entities or with the presence or absence of the ambiguity stoping the determination to entity.
It can be relatively simple that some entities determine.For example, classification select can have been based on multiple towards
The presence of the website of classification, and each of site document towards classification instruction identical entity.At this
In one situation, the entity from the site document towards classification can be selected as this entity.
In another example, one or more documents may be from the website towards classification, but the carrying of entity information
Take the multiple possible entities of generation.This can variously solve.One option can be selected at maximum
The entity occurring in the document towards classification of quantity.Another option can be the document selecting from maximum quantity
The entity of middle extraction, and do not consider source.The entity that this option will include defining based on open format module.
Another option can be at least partially based on the ranking of the document therefrom extracting each entity to select entity.
Other option can based on by each weight give from document extract data use, including with above option
Any one of combination.
Another example can relate to the situation that two or more classifications to be identified by Search Results.In some embodiments
In, can first determine that classification, and subsequently only consider the entity in selected classification.In another option,
Each document can be analyzed according to each potential classification.May then use that as above being used in multiple entities
Between the method that makes a distinction selecting entity.This will lead to corresponding classification to select.Note, in this class
In the embodiment of type, class weight can be included, as determine which entity be optimal for search inquiry
Another factor during coupling.
Another option can relate to make a distinction the situation requiring more than an information between entities.For example, perhaps
Many dining rooms are the local manufacturing enterprises of an only position.As a result, may have with same names more than a city
Dining room, although each dining room is different entities.Multiple dining rooms entity with same names is referred to alternatively as tool
There is the entity that identical principal mark is known.
Case above may result in multiple dining rooms review site of the comment with the dining room known based on identical principal mark,
But each comment is for different dining room entities.In this case, some url component matchings and other
The presence of metadata can explicitly indicate that restaurant category, but may need to further determine that to select to look into search
Ask corresponding entity.One option is to look at the data field of the additional extraction of classification.In the example in dining room,
Typical additive information for extracting may include address and telephone number information.These fields can be compared
To identify the different dining rooms entity of shared same names.This is referred to alternatively as the secondary information of each entity of comparison.?
After being made a distinction between each entity based on secondary information, the result corresponding to each different entities can be divided
Class becomes the group corresponding to each different entities.Method as above can be applied to determine and search inquiry phase
The entity of association, the entity of document such as pass through the selection entity occurring most frequently, selecting to have top ranked,
Or additive method.
In certain embodiments, entity analysis may result in the entity not being associated with inquiry.For example, if
Due to lacking url component matching without distributing classification, then can stop entity analysis process in this point.Make
For another option, entity can be determined using marking system, and for making distribution, there is no entity
Can have sufficiently high fraction and/or the fraction different enough from other potential entities.In above restaurant example
In, each dining room can only occur in a document.Marking system may need in more than one document
Occur, to obtain enough fractions for being assigned as entity.Or, two dining rooms may alternatively appear in quite
In the document of quantity, two dining rooms are led to have similar fraction.Because fraction is not different, therefore do not have
Entity can be assigned to search inquiry.
In other embodiment, multiple entities may be selected.In these embodiments, can more than an entity
Meet the criterion being selected as entity.For example, may be selected all identified entities, or can be chosen be higher than
The entity of the fraction of threshold value.In these embodiments, can be selected each entity extraction entity information.
Multiple selected entities may be from single classification, or also can identify multiple entity classes.For example, optional
Select the entity corresponding to book and the entity corresponding to film.Optionally, can be selected each entity
Display entity card.
Entity card is extracted
After mark entity, the letter with regard to this entity can be extracted from the document returning as Search Results
Breath.Entity card can be generated using the information extracted.Entity card allows to be shown with regard to the information being intended to entity
It is shown as a part for results page, and further click on without user or other actions are to find this information.
In at least one of Search Results with the corresponding each embodiment of website towards classification, can use suitable
When class template to extract the information of entity card.The type of the information extracted can be changed based on classification.
The example of the information that can be extracted includes positional information, contact details and is directed to given entity type generally quilt
The other information of request.For example, the entity card of film may include footage, director names and film
Whether it is comedy, drama or another type of film.Restaurant entity card may include food type and price model
That encloses generally indicates.Entity card with regard to sports team may include match and the previous games of next arrangement
Result.
The another type of information in entity card that may include be to one of other kinds of related content or
Multiple links.In certain embodiments, the additional information being presented in entity card may correspond to and search inquiry
The related information of secondary intent.For example, the search inquiry related to the film of current cinema projection is possible to
The result such as film comment and theater location is provided.On the contrary, in movie theatre, the film of no longer projection would be possible to
There is the result related to the shop of the copy of this film commercially available.This difference of the type of Search Results can
Represent the difference of the secondary intent of search inquiry.Can be included using this secondary intent information and secondary intent phase
The link closed is as a part for entity card.Including the link in entity card can with as from search engine
A part for result link corresponding or do not correspond to.The property of additional links may depend on entity and changes.
For dining room, link can be supplied to and process the online website subscribed.For physical culture such as film or bands
Or entertainment entity, link can be supplied to the website with obtainable admission ticket.Also link can be supplied to
Know one or more third party's review site of the comment for processing the category.
Forming one of advantage of entity card based on Search Results is that information can be dynamically generated.By
This, any change of the information being reflected in Search Results also can be automatically updated in entity card.However,
The entity card of dynamic construction can be used in combination with the static entities card of information comprising previously to have obtained.Cannot be from
In the case of extracting information needed in Search Results, it is helpful to using the previous information obtaining.
In another embodiment, entity can be identified, and the entity card including the information being stored can be provided.
In this embodiment, can identify and select entity using entity identification method described above.Subsequently may be used
Form entity card using the information being stored corresponding to selected entity.
Placed based on the information of entity detection
Can be changed using the intention of the Search Results related to entity to result and associated information
Place and/or show.After determining the intention entity of search inquiry, result can be checked with mark and entity
Relevant any result.These may include result corresponding to the website towards classification, include identified reality
The additional information that the result of the title of body or the entity with regard to being identified are not successfully extracted.
The mark of entity can variously carry out the placement of modification information.In one embodiment, the mark of entity
Know the selection that may result in the advertisement related to entity.Selected advertisement can be placed on the page close to right
Should be in the position of the Search Results of entity.For example, if corresponding to the searching of the top ranked of the entity being identified
Hitch fruit is seven to nine, then can place an advertisement on the bottom close to the page illustrating front ten Search Results.
Similarly, if generating entity card, entity card can be placed on searching of the top ranked related to entity
Hitch fruit is nearby or close to the be number two high result related to entity.
Entity detects that another impact can be to remove some items from the display of Search Results.For example, can be by
One or more documents from Search Results are merged into entity card.Optionally, can be from shown search
These results are removed, because alternatively can conduct interviews to these documents via entity card in the results list.Repair
Changing the another way that result shows can be the part showing response results, such as only display and entity or with
The related response results of the classification of entity.In this embodiment, once having carried out classification and the distribution of entity,
Can be omitted from result shows with classification and/or the unmatched result of entity.On the contrary, can be displayed in additional use
User is allowed to access the object of excluded result after the action of family.For example, it is possible to provide link is more to indicate
Incoherent result is available with the entity being identified.This link can be clicked through by user or by referring to
Pin or cursor move to access above link position.In addition, it is possible to provide there is the drop-down menu of additional result.
The example of entity detection
In this hypothetical example, user keys in search terms " godfather " in search engine first.By this
The result that search generates includes the multiple results from least one website towards classification related to film.
The additional result towards classification related to retail sales and/or video-game also can be in Search Results.Cause
Be towards classification website be top ranked Search Results, so classification selects to be face based on top ranked
To make to the website of classification.As a result, classification " film " is chosen.
After selecting classification, detect entity using the website towards classification.This leads to multiple realities is detected
Body, because film " godfather " and film " godfather ii " are both included in Search Results.Based on
" godfather ii " compares, and the fact that " godfather " is detected in more response results, film " godfather "
It is selected as suitable entity.Subsequently response results are presented to use together with the entity card corresponding to this film
Family.Entity card is to be formed based on the information extracted in the document listed from response results.
After checking presented result, search terms is revised as " godfather dining room " by user.One group new
Search Results are generated.In new result, this general comment corresponding with general review site the most front of grading
Website can be towards classification but for many classifications.Many additional potential website quilts towards classification
Including in front 20 results, corresponding to review site known to other.Based on from review site document
Metatag, have selected classification " dining room ".
Selected based on the category, suitable class template may be selected to analyze this two comment stations towards classification
Point.It is also possible to use open form class template to analyze other documents.Search Results are included around the U.S.
Some difference dining rooms and chain Pizza dining room.However, uniquely repeating of position data is to be directed to
Santiago of California.The document listing Santiago address of California is grouped in
Together, and this entity is selected as the entity corresponding to search inquiry.Note, if each example in dining room
Only occur in that once then do not have the entity will be identified in certain embodiments, can be unclear because being intended to.With
The additional information with regard to this entity can be extracted afterwards from the response results corresponding to entity.
After the general view schematically illustrating various embodiments of the present invention, presently describe and be adapted for carrying out showing of the present invention
Example property operating environment.Briefly refer to the attached drawing, and first with specific reference to Fig. 1, show for realizing this
The Illustrative Operating Environment of bright each embodiment, and it is designated generally as computing device 100.Calculating sets
Standby 100 is an example of suitable computing environment, and is not intended to the range to the present invention or work(
Any restriction can be proposed.Also computing device 100 should not be construed to shown any component or its group
Conjunction has any dependency or requirement.
Various embodiments of the present invention can set by computer or such as personal digital assistant or other hand-held
Standby etc the computer code of other machines execution or machine can using instruction (inclusion such as program module it
The computer executable instructions of class) general context described in.In general, include routine, program,
The program module of object, assembly, data structure etc. refers to execute particular task or realizes specific abstract data
The code of type.The present invention can be implemented in various system configurations, and these system configurations include hand-held and set
Standby, consumption electronic product, general purpose computer, more dedicated computing equipment etc..The present invention can also be wherein
Task is implemented by the distributed computing environment by the remote processing devices execution of communication network links.
With continued reference to Fig. 1, computing device 100 includes the bus 110 directly or indirectly coupling following equipment:
Memorizer 112, one or more processors 114, one or more assume assembly 116, input/output (i/o)
Port 118, i/o assembly 120 and illustrative power supply 122.Bus 110 can represent one or more bus
(such as address bus, data/address bus or a combination thereof).Although for the sake of clarity being shown using lines
Each frame of Fig. 1, but in fact, the profile of each assembly is not clear like that, and metaphorically for,
Lines more accurately will be Lycoperdon polymorphum Vitt and obscure.It may for instance be considered that such as display device assume assembly
It is i/o assembly.In addition, many processors have memorizer.With regard to this point, inventor recognizes that this is ability
The characteristic in domain, and reaffirm, the diagram of Fig. 1 simply can make in conjunction with one or more embodiments of the invention
The illustration of exemplary computer device.Such as " work station ", " server ", " laptop computer ",
As broad as long between the classification such as " portable equipment ", all of which is considered as within the scope of Fig. 1
And be referred to as " computing device ".
Computing device 100 generally includes various computer-readable mediums.Computer-readable medium can be can be by
Any usable medium that computing device 100 accesses, and comprise volatibility and non-volatile media, may move
With irremovable medium.Unrestricted as an example, computer-readable medium can include Computer Storage and be situated between
Matter and communication media.Computer-readable storage medium is included for storage such as computer-readable instruction, data knot
Volatibility that any method or technique of the information of structure, program module etc or other data is realized and non-volatile
Property, removable and irremovable medium.Computer-readable storage medium includes, but not limited to random access memory
Device (ram), read only memory (rom), Electrically Erasable Read Only Memory (eeprom),
Flash memory or other memory technology, cd-rom, digital versatile disc (dvd) or other holographic memory,
Cartridge, tape, disk storage or other magnetic storage apparatus, carrier wave or can be used for encode information needed and
Any other medium that can be accessed by computing device 100.In one embodiment, computer-readable medium
It can be tangible computer computer-readable recording medium.In another embodiment, computer-readable medium can be non-transient
Computer-readable medium.
Memorizer 112 includes volatibility and/or the computer-readable storage medium of nonvolatile memory form.Deposit
Reservoir can be moveable, immovable, or a combination of both.Exemplary hardware devices include solid-state
Memorizer, hard disk drive, CD drive etc..Computing device 100 is included from such as memorizer 112
Or the various entities of i/o assembly 120 etc read the one or more processors of data.Assume assembly 116
Assume data instruction to user or other equipment.Exemplary presentation components include display device, speaker, beat
Print assembly, vibration component etc..
I/o port 118 can allow computing device 100 to be logically coupled to including i/o assembly 120
Other equipment, some in these equipment can be built-in.Illustrative components include microphone, stick,
Game mat, dish, scanner, printer, wireless device etc..
Turning now to Fig. 2, show the exemplary computer system 200 illustrating according to one embodiment of the invention
Block diagram.Those skilled in the art will appreciate that and understand, the computing system 200 shown in Fig. 2 is only
The example of one suitable computing system environment, and it is not intended to the range to various embodiments of the present invention
Or function proposes any restriction.Also computing system 200 should not be construed to any list shown in it
The combination of individual assembly or each assembly has any dependency or requirement.Additionally, computing system 200 can be used as independence
Product, the part as software development environment or its any combinations are providing.
Computing system 200 includes user equipment 206 and the search service 208 communicating with one another via network 204.
Search service 208 may include search engine 212, entity identification component 214, template storage 216 and knot
Fruit assumes assembly 218.Search engine 212 could be for generating the normal of response results based on search inquiry
Rule search engine.Entity identification component 214 can analyze Search Results to determine classification and to look into corresponding to search
The entity ask.This analysis can be held partially by using the class template being stored in template storage 216
OK.Result assumes assembly 218 and can change response using the entity information being provided by entity identification component 214
The display of result.Based on the entity being identified, can by the advertisement of the mark based on entity include with regard to institute
The corresponding position of result of the entity of mark.It is also based on identified entity to assume entity card.
Fig. 3 depicts the flow chart illustrating the method according to one embodiment of the invention.Reality shown in Fig. 3
Apply in example, obtain the multiple results 310 in response to search inquiry.Result can be obtained from remote search engines,
Or result based on receiving search inquiry and can generate one group of response results.Can be by one or more response results
Carry out mating 320 with the class template corresponding to one or more websites towards classification.In embodiments,
Response results can all be mated with a class template, or result can be with the mould of all categories from single entity classification
Plate mates, or result can be mated with the class template from multiple entity classes.Based on one or more couplings
Response results selecting entity class 330.Also it is based on response results and identify entity 340.In each enforcement
In example, entity 340 can be identified after selecting entity class 330.Subsequently retrieval is corresponding to the reality being identified
The entity information 350 of the storage of body.The entity information retrieved is merged into together with least one response results
Entity card 360.Display does not include the response results list 370 of merged response results.Entity card also by
Display 380.Note, in this embodiment, based on being merged into shown entity card, the response being merged
Result is excluded beyond response results list.
Fig. 4 depicts the flow chart illustrating the method according to one embodiment of the invention.Reality shown in Fig. 4
Apply in example, acquisition 410 is in response to multiple results of search inquiry.Result can be obtained from remote search engines,
Or result based on receiving search inquiry and can generate one group of response results.Can be by one or more response results
Carry out mating 420 with the class template corresponding to one or more websites towards classification.In embodiments,
Response results can all be mated with a class template, or result can be with the mould of all categories from single entity classification
Plate mates, or result can be mated with the class template from multiple entity classes.Based on one or more couplings
Response results selecting entity class 430.Identify multiple entities 440 based on response results.In each reality
Apply in example, multiple entities 440 can be identified after selecting entity class 430.Multiple entities each have phase
Same principal mark is known.The secondary information 450 of each of multiple entities is extracted from response results.Based on institute
The secondary information extracted is by the entity classification being identified to each group of entities 460.This classification can allow to be phase
The entity being identified with entity is grouped together, and each different entities have separate group.Subsequently may be used
Select group of entities 470 corresponding with search inquiry.For selected group of entities it may be determined that corresponding to this in fact
The response results of body group.The secondary information of selected group of entities is merged into 490 in entity card.Display is real
A part 495 for body card and response results.A part for response results can be excluded and is determined to correspond to reality
The response results of body group.
Fig. 5 depicts the flow chart illustrating the method according to one embodiment of the invention.Reality shown in Fig. 5
Apply in example, obtain the multiple results 510 in response to search inquiry.Result can be obtained from remote search engines,
Or result based on receiving search inquiry and can generate one group of response results.Can be by one or more response results
Carry out mating 520 with the class template corresponding to one or more websites towards classification.In embodiments,
Response results can all be mated with a class template, or result can be with the mould of all categories from single entity classification
Plate mates, or result can be mated with the class template from multiple entity classes.Based on one or more couplings
Response results selecting entity class 530.Also it is based on response results and identify entity 540.In each enforcement
In example, entity 540 can be identified after selecting entity class 530.Subsequently can extract real from response results
Body information 550.The entity information being extracted of each of each entity entity can be merged into entity card
560.Multiple entity card 570 can be shown.
Additional embodiment
In one embodiment, there is provided storage computer can be using one or more Computer Storage of instruction
Medium, when these instructions are executed by computer equipment, is executed for determining the reality being associated with search inquiry
The method of body.The method includes obtaining the multiple results in response to search inquiry.Multiple response results are mated
To the multiple class templates corresponding to the website towards classification.Can based on the plurality of mated response results Lai
Select entity class.Identify entity based on multiple mated response results.Retrieval is with regard to the reality being identified
The entity information of the storage of body.The entity information retrieved and at least one response results are merged into entity
Card.Show response results in search result list, this list does not include this at least one response of being merged
Result.Entity card is also shown.
In another embodiment, there is provided storage computer can be deposited using one or more computers of instruction
Storage media, when these instructions are executed by computer equipment, is executed and is associated with search inquiry for determining
The method of entity.The method includes obtaining the multiple results in response to search inquiry.By one or more responses
Result matches at least one class template corresponding to the website towards classification.Based on one or more couplings
Response results selecting entity class.Mark is corresponding to the entity class selected in multiple response results
Multiple entities, each entity being identified has identical principal mark to be known.Extract from multiple response results and correspond to
Secondary identification information in each entity being identified.To be identified based on the secondary identification information being extracted
Entity classification become one or more group of entities.Group of entities is elected as corresponding with search inquiry.Determine corresponding to
One or more response results of selected group of entities.The secondary mark being extracted by selected group of entities
Knowledge information is merged in entity card.Display entity card and a part for response results, this portion of response results
Divide one or more response results of the entity corresponding to selected group of entities determined by eliminating.
In another embodiment, there is provided the method for determining the entity being associated with search inquiry.The party
Method includes obtaining the multiple results in response to search inquiry.By one or more response results match corresponding to
At least one class template towards the website of classification.Response results based on one or more couplings are selecting
One or more entity classes.Identify multiple entities based on multiple response results.From multiple response results
Extract the entity information corresponding to multiple identified entities.By each of multiple identified entities
The entity information being extracted is merged in multiple entity card.Show multiple entity card.
Describe various embodiments of the present invention with reference to each specific embodiment, each specific embodiment is in all respects all
It is intended to be illustrative and be not restrictive.In the case of the scope of the invention, each alternative embodiment pair
Will become clear from those skilled in the art in the invention.
It will be observed from the foregoing that the present invention is perfectly suitable for realizing all purposes described above
And target, and to have for this structure be other apparent and intrinsic advantages.It will be understood that it is specific
Feature and sub-portfolio are useful, and can be employed without with reference to other features and sub-portfolio.This
Conceived by claim, and within the scope of the claims.
Claims (15)
1. a kind of method for determining the entity being associated with search inquiry, methods described includes:
Obtain the multiple results (310) in response to search inquiry；
The response results of the plurality of result are matched the multiple classification moulds corresponding to the website towards classification
Plate (320), described class template includes assuming form for the corresponding website towards classification, described
Website towards classification is defined as following the tracks of the website of the state of the multiple entities in entity class；
Distributed selected from multiple entity classes based on the response results mating the plurality of class template
Entity class (330)；
Extract at least based on the entity class being distributed belonging to entity and from the response results mated
One data word section is come to identify entity (340), at least one data field described be using the plurality of class
Come to extract, at least one data field described is different from url assembly to the extraction format component of other template；
The entity information (350) of the entity with regard to being identified of retrieval storage；
The entity information retrieved and at least one response results are merged in entity card (360)；
Show described response results (370) in search result list, described search result list does not include
At least one response results being merged；And
Show described entity card (380)；
Wherein allocation entity classification includes:
The quantity next life of the response results based on the specific class template mating the plurality of class template
Become the classification fraction of multiple entity classes；And
Distribute described entity class based on the highest classification fraction that has of the plurality of entity class.
2. the method for claim 1 it is characterised in that multiple response results are matched right
Ying Yu includes a part for the URL of document towards multiple class templates of the website of classification
Match with the URL assembly of class template.
3. the method for claim 1 is it is characterised in that also include with carrying from response results
The additional entities information taking is supplementing stored entity information.
4. the method for claim 1 is it is characterised in that merge at least one response results bag
Include at least one mated response results of merging.
5. the method for claim 1 is it is characterised in that mark entity includes:
Generated many based at least one data field extracting from one or more of response results
The entity score of each entity of individual entity；And
Mark has the entity of highest entity score.
6. a kind of method for determining the entity being associated with search inquiry, methods described includes:
Obtain (410) multiple results in response to search inquiry；
One or more response results of multiple results are matched corresponding to the website towards classification at least
One class template (420), the described website towards classification is defined as multiple in tracking entity class
The website of the state of entity；
Based on one or more response results of coupling at least one class template described, entity class is divided
Search inquiry described in dispensing (430)；
Identify the multiple entities (440) corresponding to the entity class being distributed in the plurality of response results,
Each identified entity is based on belonging to described entity class and extract from the response results mated
At least one data field is identified, and at least one data field described is using at least one classification mould described
The extraction format component of plate is extracted, and at least one data field described is different from url assembly；
The secondary identification information corresponding to each identified entity is extracted from the plurality of response results
(450)；
Based on the secondary identification information being extracted come by the multiple entity classifications being identified to one or more realities
Body group (460)；
Group of entities is elected to be corresponding with described search inquiry (470)；
Determine the one or more response results (480) corresponding to selected group of entities；
The secondary identification information being extracted of selected group of entities is merged in entity card (490)；With
And
Show a part (495) for described entity card and described response results, described response results should
One or more response results of the entity of selected group of entities are corresponded to determined by partly eliminating.
7. method as claimed in claim 6 is it is characterised in that determine corresponding to selected entity
One or more response results of the entity of group include mark from the secondary wherein extracting selected group of entities
The response results of identification information.
8. method as claimed in claim 6 is it is characterised in that by one or more response results
It is fitted at least one class template corresponding to the website towards classification to include the unified resource positioning of document
A part for symbol is matched with the URL assembly of class template.
9. method as claimed in claim 6 is it is characterised in that also include:
By what selected group of entities mark was extracted from multiple response results, there is being extracted of conflict
At least one secondary identification information field of value；
The value being stored is distributed to the letter of the value extracted in the entity information being extracted with conflict
Breath field.
10. method as claimed in claim 6 is it is characterised in that select group of entities to include from highest number
Select in the response results of amount to comprise the group of entities of extracted secondary information and/or from least multiple documents
Middle selection comprises the group of entities of extracted secondary information.
11. methods as claimed in claim 6 are it is characterised in that tie from one or more of responses
Extract secondary information in fruit to include extracting number from one or more documents based on opening mode class template
According to field.
A kind of 12. methods for determining the entity being associated with search inquiry, comprising:
Obtain the multiple results (510) in response to search inquiry；
By the document matches from described result at least one the classification mould corresponding to the website towards classification
Plate (520), at least one class template described includes assuming lattice for the corresponding website towards classification
Formula, the described website towards classification is defined as following the tracks of the website of the state of multiple entities in entity class；
Generate the classification fraction of multiple entity classes based on described result, be wherein based at least one described in coupling
The quantity of multiple documents of class template is giving an extra weight of classification fraction；
Based on described classification fraction by one or more entity classes of multiple entity classes distribute to described in search
Rope is inquired about；
Based on the one or more entity classes at least being distributed and from the document of one or more couplings
At least one data field extracting identifying multiple entities (540), at least one data field described is
It is extracted using the extraction format component of at least one class template described, at least one data field described
Different from url assembly；
The entity information (550) corresponding to multiple identified entities is extracted from multiple response results；
Based on described search inquiry by the entity information being extracted of each of multiple identified entities
It is merged into (560) in multiple entity card；And
Show the plurality of entity card (570).
13. methods as claimed in claim 12 are it is characterised in that also include exclusion at least one
The display of unmatched response results.
14. methods as claimed in claim 13 are it is characterised in that exclusion is at least one mismatch
Response results display include provide at least one unmatched response results compression expression, described compression
Represent and need at least one additional user action to show at least one unmatched response results described.
A kind of 15. systems for determining the entity being associated with search inquiry, described system includes:
For obtaining the device of the multiple results in response to search inquiry；
For the response results of the plurality of result being matched the multiple classes corresponding to the website towards classification
The device of other template, described class template includes the form that presents for the corresponding website towards classification, institute
State the website towards classification to be defined as following the tracks of the website of the state of the multiple entities in entity class；
For distributing institute from multiple entity classes based on the response results mating the plurality of class template
The device of the entity class of choosing；
For based on the entity class being distributed belonging to entity and from the response results extraction mated
Identifying the device of entity, at least one data field described is using the plurality of at least one data field
Come to extract, at least one data field described is different from url assembly to the extraction format component of class template；
Device for the entity information of entity with regard to being identified of retrieval storage；
For the entity information retrieved and at least one response results are merged into the device in entity card；
For showing the device of described response results in search result list, described search result list does not wrap
Include the response results that at least one is merged；And
For showing the device of described entity card；
The device being wherein used for allocation entity classification includes:
Quantity for the response results based on the specific class template mating the plurality of class template
To generate the device of the classification fraction of multiple entity classes；And
Distribute described entity class for the highest classification fraction that has based on the plurality of entity class
Device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610943349.8A CN107103018A (en) | 2010-06-10 | 2011-06-09 | The entity detection and extraction of entity card |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US12/813,390 | 2010-06-10 | ||
| US12/813,390 US9158846B2 (en) | 2010-06-10 | 2010-06-10 | Entity detection and extraction for entity cards |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610943349.8A Division CN107103018A (en) | 2010-06-10 | 2011-06-09 | The entity detection and extraction of entity card |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102393840A CN102393840A (en) | 2012-03-28 |
| CN102393840B true CN102393840B (en) | 2017-01-18 |
Family
ID=45097081
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110164789.0A Active CN102393840B (en) | 2010-06-10 | 2011-06-09 | Entity detection and extraction for entity cards |
| CN201610943349.8A Pending CN107103018A (en) | 2010-06-10 | 2011-06-09 | The entity detection and extraction of entity card |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610943349.8A Pending CN107103018A (en) | 2010-06-10 | 2011-06-09 | The entity detection and extraction of entity card |
Country Status (2)
| Country | Link |
|---|---|
| US (2) | US9158846B2 (en) |
| CN (2) | CN102393840B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120246154A1 (en) * | 2011-03-23 | 2012-09-27 | International Business Machines Corporation | Aggregating search results based on associating data instances with knowledge base entities |
| US9959326B2 (en) | 2011-03-23 | 2018-05-01 | International Business Machines Corporation | Annotating schema elements based on associating data instances with knowledge base entities |
| US8504561B2 (en) * | 2011-09-02 | 2013-08-06 | Microsoft Corporation | Using domain intent to provide more search results that correspond to a domain |
| US8856109B2 (en) * | 2012-06-21 | 2014-10-07 | Microsoft Corporation | Topical affinity badges in information retrieval |
| US9727545B1 (en) * | 2013-12-04 | 2017-08-08 | Google Inc. | Selecting textual representations for entity attribute values |
| WO2015113306A1 (en) | 2014-01-30 | 2015-08-06 | Microsoft Corporation | Entity page generation and entity related searching |
| US20150317945A1 (en) * | 2014-04-30 | 2015-11-05 | Yahoo! Inc. | Systems and methods for generating tinted glass effect for interface controls and elements |
| US9916328B1 (en) | 2014-07-11 | 2018-03-13 | Google Llc | Providing user assistance from interaction understanding |
| US9965559B2 (en) | 2014-08-21 | 2018-05-08 | Google Llc | Providing automatic actions for mobile onscreen content |
| US20160092082A1 (en) * | 2014-09-29 | 2016-03-31 | Apple Inc. | Visualizing Relationships Between Entities in Content Items |
| US9965474B2 (en) | 2014-10-02 | 2018-05-08 | Google Llc | Dynamic summary generator |
| US20160292282A1 (en) * | 2015-04-01 | 2016-10-06 | Microsoft Technology Licensing, Llc | Detecting and responding to single entity intent queries |
| US9703541B2 (en) | 2015-04-28 | 2017-07-11 | Google Inc. | Entity action suggestion on a mobile device |
| US10970646B2 (en) | 2015-10-01 | 2021-04-06 | Google Llc | Action suggestions for user-selected content |
| US10178527B2 (en) | 2015-10-22 | 2019-01-08 | Google Llc | Personalized entity repository |
| US10055390B2 (en) | 2015-11-18 | 2018-08-21 | Google Llc | Simulated hyperlinks on a mobile device based on user intent and a centered selection of text |
| US10535005B1 (en) | 2016-10-26 | 2020-01-14 | Google Llc | Providing contextual actions for mobile onscreen content |
| US10824630B2 (en) * | 2016-10-26 | 2020-11-03 | Google Llc | Search and retrieval of structured information cards |
| US11237696B2 (en) | 2016-12-19 | 2022-02-01 | Google Llc | Smart assist for repeated actions |
| US11586820B2 (en) * | 2020-08-10 | 2023-02-21 | Ebay Inc. | Techniques for enhancing the quality of human annotation |
| US11467716B1 (en) | 2022-01-28 | 2022-10-11 | Microsoft Technology Licensing, Llc | Flexibly identifying and playing media content from any webpage |
Family Cites Families (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6144958A (en) | 1998-07-15 | 2000-11-07 | Amazon.Com, Inc. | System and method for correcting spelling errors in search queries |
| US6735585B1 (en) * | 1998-08-17 | 2004-05-11 | Altavista Company | Method for search engine generating supplemented search not included in conventional search result identifying entity data related to portion of located web page |
| US7096218B2 (en) | 2002-01-14 | 2006-08-22 | International Business Machines Corporation | Search refinement graphical user interface |
| US6978268B2 (en) | 2002-03-16 | 2005-12-20 | Siemens Medical Solutions Health Services Corporation | Healthcare organization central record and record identifier management system |
| US20040049514A1 (en) * | 2002-09-11 | 2004-03-11 | Sergei Burkov | System and method of searching data utilizing automatic categorization |
| US7613687B2 (en) * | 2003-05-30 | 2009-11-03 | Truelocal Inc. | Systems and methods for enhancing web-based searching |
| US7536382B2 (en) | 2004-03-31 | 2009-05-19 | Google Inc. | Query rewriting with entity detection |
| US7949642B2 (en) | 2004-10-12 | 2011-05-24 | Wendy W Yang | System and method for managing and presenting entity information |
| US8719244B1 (en) * | 2005-03-23 | 2014-05-06 | Google Inc. | Methods and systems for retrieval of information items and associated sentence fragments |
| US20070100650A1 (en) | 2005-09-14 | 2007-05-03 | Jorey Ramer | Action functionality for mobile content search results |
| US20100121705A1 (en) * | 2005-11-14 | 2010-05-13 | Jumptap, Inc. | Presentation of Sponsored Content Based on Device Characteristics |
| US9135238B2 (en) | 2006-03-31 | 2015-09-15 | Google Inc. | Disambiguation of named entities |
| US7822762B2 (en) | 2006-06-28 | 2010-10-26 | Microsoft Corporation | Entity-specific search model |
| US7571158B2 (en) | 2006-08-25 | 2009-08-04 | Oracle International Corporation | Updating content index for content searches on networks |
| US7685201B2 (en) | 2006-09-08 | 2010-03-23 | Microsoft Corporation | Person disambiguation using name entity extraction-based clustering |
| US7698259B2 (en) | 2006-11-22 | 2010-04-13 | Sap Ag | Semantic search in a database |
| US7555478B2 (en) * | 2006-12-05 | 2009-06-30 | Yahoo! Inc. | Search results presented as visually illustrative concepts |
| US20080222105A1 (en) | 2007-03-09 | 2008-09-11 | Joseph Matheny | Entity recommendation system using restricted information tagged to selected entities |
| US7917489B2 (en) * | 2007-03-14 | 2011-03-29 | Yahoo! Inc. | Implicit name searching |
| US7698261B1 (en) | 2007-03-30 | 2010-04-13 | A9.Com, Inc. | Dynamic selection and ordering of search categories based on relevancy information |
| US20080263022A1 (en) * | 2007-04-19 | 2008-10-23 | Blueshift Innovations, Inc. | System and method for searching and displaying text-based information contained within documents on a database |
| GB2462564A (en) | 2007-06-28 | 2010-02-17 | Taptu Ltd | Interactive web scraping of onine content for search and display on mobile dev ices |
| KR20090012467A (en) | 2007-07-30 | 2009-02-04 | 한국과학기술정보연구원 | System and method for providing integrated search using uniform resource identifier database |
| US7836046B2 (en) * | 2008-01-21 | 2010-11-16 | Oracle Financial Services Software Limited | Method and system for facilitating verification of an entity based on business requirements |
| CN101494617B (en) | 2008-01-23 | 2010-12-15 | 华为技术有限公司 | Method, system and device for classifying content |
| US20090192983A1 (en) | 2008-01-28 | 2009-07-30 | Yahoo! Inc. | Method and system for mining, ranking and visualizing lexically similar search queries for advertisers |
| US8135707B2 (en) | 2008-03-27 | 2012-03-13 | Yahoo! Inc. | Using embedded metadata to improve search result presentation |
| US20090327223A1 (en) | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Query-driven web portals |
| GB2472358A (en) | 2008-07-02 | 2011-02-02 | Lexisnexis Risk & Information Analytics Group Inc | Database system and methods |
| US20100121842A1 (en) | 2008-11-13 | 2010-05-13 | Dennis Klinkott | Method, apparatus and computer program product for presenting categorized search results |
| US8458171B2 (en) | 2009-01-30 | 2013-06-04 | Google Inc. | Identifying query aspects |
-
2010
- 2010-06-10 US US12/813,390 patent/US9158846B2/en active Active
-
2011
- 2011-06-09 CN CN201110164789.0A patent/CN102393840B/en active Active
- 2011-06-09 CN CN201610943349.8A patent/CN107103018A/en active Pending
-
2015
- 2015-10-09 US US14/879,959 patent/US20160034471A1/en not_active Abandoned
Also Published As
| Publication number | Publication date |
|---|---|
| US20110307483A1 (en) | 2011-12-15 |
| CN102393840A (en) | 2012-03-28 |
| US9158846B2 (en) | 2015-10-13 |
| CN107103018A (en) | 2017-08-29 |
| US20160034471A1 (en) | 2016-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102393840B (en) | Entity detection and extraction for entity cards | |
| CN102279872B (en) | Inquiring intention identification drived by search results | |
| Liu et al. | Tiara: Interactive, topic-based visual text summarization and analysis | |
| US10031925B2 (en) | Method and system of using image recognition and geolocation signal analysis in the construction of a social media user identity graph | |
| CN101292238B (en) | Method and system for automated rich presentation of a semantic topic | |
| US8706674B2 (en) | Media tag recommendation technologies | |
| US9600826B2 (en) | Local metric learning for tag recommendation in social networks using indexing | |
| US20210089570A1 (en) | Methods and apparatus for assessing candidates for visual roles | |
| US7752208B2 (en) | Method and system for detection of authors | |
| US11609920B2 (en) | Ranking and presenting search engine results based on category-specific ranking models | |
| US20120303637A1 (en) | Automatic wod-cloud generation | |
| CN102289459A (en) | Automatically generating training data | |
| Liu et al. | Background music recommendation based on latent factors and moods | |
| US20120046937A1 (en) | Semantic classification of variable data campaign information | |
| Zhang et al. | Image clustering: An unsupervised approach to categorize visual data in social science research | |
| Nguyen et al. | On predicting religion labels in microblogging networks | |
| Burgess et al. | Critical simulation as hybrid digital method for exploring the data operations and vernacular cultures of visual social media platforms | |
| Dashrath | Role of metadata in digital resource management | |
| Kuhn et al. | The VAT: enhanced video analysis | |
| ALHawari et al. | The impact of knowledge process about customer on the success of customer knowledge acquisition | |
| US20140280149A1 (en) | Method and system for content aggregation utilizing contextual indexing | |
| US20190050467A1 (en) | Method and System for Content Creation and Management | |
| Lu et al. | Exploratory product image search with circle-to-search interaction | |
| Arnold et al. | Semi-Automatic Identification of Counterfeit Offers in Online Shopping Platforms | |
| US20230401570A1 (en) | Determining an optimum quantity of fractional non-fungible tokens to generate for content and a content exchange |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: MICROSOFT TECHNOLOGY LICENSING LLC Free format text: FORMER OWNER: MICROSOFT CORP. Effective date: 20150803 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20150803 Address after: Washington State Applicant after: Micro soft technique license Co., Ltd Address before: Washington State Applicant before: Microsoft Corp. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |